Using Metrics for Code Smells of ML Pipelines

被引：0

作者：

Costal, Dolors ^{[1
]}

Gomez, Cristina ^{[1
]}

del Rey, Santiago ^{[1
]}

Martinez-Fernandez, Silverio ^{[1
]}

机构：

[1] Univ Politecn Cataluna, Barcelona, Spain

来源：

IEEE 21ST INTERNATIONAL CONFERENCE ON SOFTWARE ARCHITECTURE COMPANION, ICSA-C 2024 | 2024年

关键词：

ML Pipelines; Code smells; Metrics;

D O I：

10.1109/ICSA-C63560.2024.00055

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

ML pipelines, as key components of ML systems, shall be developed following quality assurance techniques. Unfortunately, it is often the case in which they present maintainability issues, due to the experimental nature of data collection and ML model construction. In this work, we perform a first evaluation of a set of metrics, proposed in previous research, for measuring the presence of code smells related to maintainability, in ML pipeline application examples. Moreover, we provide the lessons learnt and insights gained from this evaluation.

引用

页码：289 / 294

页数：6

共 14 条

[1] Balatsko M, 2020, Your ml code is well structured... or is it?
[2] Costal D., 2024, LNCS, V14484
[3] github, TextMood project
[4] github, TeamBeans project
[5] github, WhereIs Wally project
[6] ISO, 2019, Continuous delivery for machine learning.
[7] ISO, 2017, Iso/iec 25010:2011
[8] Teaching MLOps in Higher Education through Project-Based Learning
Lanubile, Filippo
Martinez-Fernandez, Silverio
Quaranta, Luigi
[J]. 2023 IEEE/ACM 45TH INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING-SOFTWARE ENGINEERING EDUCATION AND TRAINING, ICSE-SEET, 2023, : 95 - 100
[9] Software Architecture Challenges for ML Systems
Lewis, Grace A.
Ozkaya, Ipek
Xu, Xiwei
[J]. 2021 IEEE INTERNATIONAL CONFERENCE ON SOFTWARE MAINTENANCE AND EVOLUTION (ICSME 2021), 2021, : 634 - 638
[10] Martin R.C., 2003, AGILE SOFTWARE DEV P, DOI [10.5555/515230, DOI 10.5555/515230]

← 1 2 →